The left lung of a dog was suspended in a Lucite box, ventilated with negative pressure, and perfused with venous blood from another dog at 37 C. By varying the pulmonary artery pressure, it was possible to perfuse all the lung or leave increasing proportions of the upper and middle lobes unperfused. The mean Pco2 difference between end-tidal gas and pulmonary venous blood was 1.7 mm Hg (se mean 0.6) when all the lung was perfused, but steadily increased up to 17 mm Hg as more and more of the lung was unperfused. Alveolar dead space/alveolar tidal volume increased linearly with the proportion of lung unperfused. By contrast, the venous admixture component remained small as increasing amounts of lung were unperfused. The results were compared with calculations made on a theoretical model of the lung and it was concluded that the uneven distribution of blood flow caused by hydrostatic pressure differences down the lung may seriously interfere with CO2 exchange when the pulmonary artery pressure is low, but that this type of uneven distribution affects O2 exchange much less.
isolated lung; hydrostatic effect; alveolar-arterial differences; physiologic dead space; alveolar dead space; shunt effect; venous admixture; increased acceleration; hypoxemia; hemorrhage; positive-pressure breathing
Submitted on September 10, 1964